We made significant progress towards both of our long term goals: 1) development of a gene therapy for intravascular thrombosis through high level expression of plasminogen activators from endothelial cells, and 2) the development of a gene therapy for recurrent arterial stenosis. Previous experiments demonstrated that overexpression of tissue plasminogen activator (t-PA) from endothelial cells resulted in enhanced fibrinolytic activity, but with moderation of the effect due to plasminogen activator inhibitor (PAI-1) binding. Further experiments with plasminogen activators which are resistant to PAI-1 inhibition revealed substantial residual binding by PAI-1. We therefore constructed a vector which expresses single chain urokinase (a proenzyme plasminogen activator which does not bind to PAI-1) and anchored it to the apical surface of endothelial cells with a glycolipid anchor. The resulting molecule significantly enhances cell surface fibrinolysis above that achieved with wild type plasminogen activators. We have begun testing the fibrinolytic activity of endothelial cells transduced to overexpress various plasminogen activators in an in vivo baboon model of intravascular thrombosis. Current methods of direct in vivo vascular gene transfer are extremely inefficient and must be improved before one may consider their therapeutic application. We developed a catheter-based delivery system using small, labeled particles as tracers to understand the mechanics of vector delivery into the vessel wall. We also completed an extensive study of the optimization of retrovirus-mediated gene transfer in vitro. A protocol was developed which allows transduction of 50-90% of endothelial cells with one exposure to vector. Further improvements of this protocol may allow high efficiency in vivo vascular gene transfer.